Run MPSoCBench multi_ application

[erwan....@gmail.com]

Hello Liana,

First of all, thanks for the help previously done with MPSoCBench.
MPSoCBench is working perfectly when I run applications like fft, dijkstra, stringsearch etc... However it cannot compile the rundir for "multi_" applications. The problem seems to be located at the .x generation for the interconnection, I choosed -i=noc.at
Below this text is an example of output error :

erwan@lo-lester-040:~/Documents/Logiciels/MPSoCBench/MPSoCBench$ ./MPSoCBench -p=mips -i=noc.at -n=8 -s=multi_8 -b
[['mips'], ['8'], ['multi_8'], ['noc.at']]
rm -f main.o arm_arch.o arm_arch_ref.o arm.o arm_syscall.o arm_intr_handlers.o *~ arm.x core *.o *.a
rm -f arm_arch.cpp arm_arch_ref.cpp arm.cpp arm_parms.H arm_arch.H arm_arch_ref.H arm_isa.H arm_bhv_macros.H  arm_intr_handlers.H arm_ih_bhv_macros.H arm.H arm_isa_init.cpp *.tmpl loader.ac
rm -f main.cpp Makefile
make[1]: *** Pas de règle pour fabriquer la cible « clean ». Arrêt.
make[1]: *** Pas de règle pour fabriquer la cible « distclean ». Arrêt.
Making Processor mips ...
ArchC: Parsing AC_ARCH declaration file: mips_nonblock.ac
ArchC: Parsing AC_ISA declaration file: mips_isa.ac
ArchC: mips model files generated.
ar: création de libmips.a
Making IP tlm_memory_at ...
ar: création de libtlm_memory_at.a
Making IP tlm_lock_at ...
ar: création de libtlm_lock_at.a
Making IP tlm_dfs_at ...
ar: création de libtlm_dfs_at.a
Making IP tlm_intr_ctrl_at ...
ar: création de libtlm_intr_ctrl_at.a
Making IS tlm_noc_at ...
ar: création de libtlm_noc_at.a
Making Software multi_8 ...
make[1]: *** Pas de cible spécifiée et aucun makefile n'a été trouvé. Arrêt.
cp: impossible d'évaluer «sw/multi_8/*.x»: Aucun fichier ou dossier de ce type
make: *** [all] Erreur 1
Creating rundir for mips.noc.at.8.multi_8...
---copy platform.noc.at.x to the appropriate rundir
cp: impossible d'évaluer «platform.noc.at.x»: Aucun fichier ou dossier de ce type
make[1]: *** [copy] Erreur 1
make: *** [copy] Erreur 2
make[1]: *** Pas de règle pour fabriquer la cible « clean ». Arrêt.
erwan@lo-lester-040:~/Documents/Logiciels/MPSoCBench/MPSoCBench$


In fact, I have the same error for each soft with the word "multi" included in the name. Do you have an idea how I can solve this issue ?
Thank you in advance for your support.

Best regards,
Erwan

[Liana Duenha]

Hello Erwan,

I'll check and answer you then. 

I think the problem is with the character "_" (and not with the "multi").

I need to know which version of MPSoCBench and ArchC you are using. 

Liana.

--
You received this message because you are registered to the group "MPSoCBench" Google Groups.
To post to this group, send email to mpsoc...@googlegroups.com
Visit this group http://groups.google.com/group/mpsocbench?hl=en
---
You received this message because you are subscribed to the Google Groups "MPSoCBench" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mpsocbench+...@googlegroups.com.
To post to this group, send email to mpsoc...@googlegroups.com.
Visit this group at https://groups.google.com/group/mpsocbench.

--

-------------------------------------------------------------------------
Liana Duenha

Faculdade de Computação - FACOM/UFMS

Website Institucional: www.facom.ufms.br

Website Pessoal: www.facom.ufms.br/~lianaduenha

e-mail: liana...@facom.ufms.br
           liana...@gmail.com

Fone: +55 67 3345-7846

Faculty of Computing - FACOM/UFMS

Faculty Website: www.facom.ufms.br

Personal Website: www.facom.ufms.br/~lianaduenha

e-mail: liana...@facom.ufms.br
           liana...@gmail.com

Phone: +55 67 3345-7846

[Erwan Moréac]

Hello Liana,

In the file MPSoCBench frontend script there are information about software names and to launch multi applications, the character "_" is used.

Here are the software versions of my setup :

MPSoCBench is v2.0.0
Révision: 135b7f8
Auteur: maxiwell
Date: 2015-10-22 19:23:11
setup.sh working with offline processors
"If the models are present, the setup.sh does not run the 'git submodule'"

And I use ARchC-2.4.0.

Thank you for the support.
Best regards,

Erwan

You received this message because you are subscribed to a topic in the Google Groups "MPSoCBench" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mpsocbench/EJyRyuKPnTk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mpsocbench+...@googlegroups.com.

[Liana Duenha]

Hello,

I reviewed the MPSoCBench git log and I we saw the lack of some application Makefiles in the git repository (apps with "_" in the name). 

Please, can you verify if the revision you are using contains a Makefile file in each app folder ? 

Example:  sw/multi_8/Makefile  ?

                 sw/multi_16/Makefile  ?

                 sw/water_spatial/Makefile  ?

Liana.

[Erwan Moréac]

You right, in sw/, all the multi_ applications haven't a makefile. You can check this with the picture attached to this mail.

Should I download a newer MPSoCBench version ?

Best regards,

Erwan

[Liana Duenha]

Of course Erwan!

Thank you for your collaboration!

We will make available the latest revision until the end of the day.

Liana.

[Liana Duenha]

Erwan,

MPSoCBench 2.2 available:  http://www.archc.org/benchs/mpsocbench/downloads.html

Please let us know if you have problems to download and install the new release.

Liana.

[Erwan Moréac]

Hello Liana,

Sorry for the slow reply, I had to test the multi-applications. I downloaded MPSoCBench v2.2 and built some multi apps. There is no error when the app is compiled (I just have multisoftware_16.c:478:3: warning: incompatible implicit declaration of built-in function '_exit' [enabled by default]   _exit(0);   ^), however there is a problem during the simulation. I put an example below :

erwan@lo-lester-040:~/Documents/Logiciels/MPSoCBench/MPSoCBench$ ./MPSoCBench -p=mips -i=noc.at -n=8 -s=multi_8 -r

[['mips'], ['8'], ['multi_8'], ['noc.at']]

./platform.noc.at.x multi_8.mips.x 8

        SystemC 2.3.1-Accellera --- Apr 16 2015 10:39:56
        Copyright (c) 1996-2014 by all Contributors,
        ALL RIGHTS RESERVED

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_0
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_1
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_2
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_3
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_4
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_5
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_6
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_7
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_8
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_9
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_10
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_11
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_12
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_13
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_14
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

NOC:

4 Lines

4 Columns

11 Peripherals

8 Masters

3 Slaves

5 Inactive Nodes

8 Slave Empty Nodes

3 Master Empty Nodes

16 Wrappers for Masters/Slaves

Printing mesh!

mesh[0][0]: status-> 1  x->0  y->0
mesh[0][1]: status-> 1  x->0  y->1
mesh[0][2]: status-> 1  x->0  y->2
mesh[0][3]: status-> 1  x->0  y->3
mesh[1][0]: status-> 1  x->1  y->0
mesh[1][1]: status-> 1  x->1  y->1
mesh[1][2]: status-> 1  x->1  y->2
mesh[1][3]: status-> 1  x->1  y->3
mesh[2][0]: status-> 1  x->2  y->0
mesh[2][1]: status-> 1  x->2  y->1
mesh[2][2]: status-> 1  x->2  y->2
mesh[2][3]: status-> 0  x->2  y->3
mesh[3][0]: status-> 0  x->3  y->0
mesh[3][1]: status-> 0  x->3  y->1
mesh[3][2]: status-> 0  x->3  y->2
mesh[3][3]: status-> 0  x->3  y->3
Printing Table of routs!
Number of Active Lines in the table of routs: 11
m[0]: 536870912 0 0
 m[1]: 553648128 0 1
 m[2]: 570425344 0 2
 m[3]: 587202560 0 3
 m[4]: 603979776 1 0
 m[5]: 620756992 1 1
 m[6]: 637534208 1 2
 m[7]: 654311424 1 3
 m[8]: 671088640 2 0
 m[9]: 687865856 2 1
 m[10]: 704643072 2 2
 ArchC: Reading ELF application file: multi_8.mips.x


ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

MPSoCBench: The simulator is prepared.
MPSoCBench: Beggining of time simulation measurement.

Warning: (W536) immediate self-notification ignored as of IEEE 1666-2011: noc.thread_node
In file: ../../../../src/sysc/kernel/sc_process.cpp:361
In process: noc.thread_node @ 100 ns

--------------------------------------------------------------------
-------------------------  MPSoCBench  -----------------------------
-----------------------Running: multi_8 ----------------------------
---------SHA, Rijndael Encoder, Rijndael Decoder, Blowfish----------
---------PBM_stringsearch, BMH_stringsearch, FFT, ADPCM-------------
--------------------------------------------------------------------


Stringsearch Pratt-Boyer-Moore has finished.

ArchC: -------------------- Simulation Finished --------------------
ArchC Runtime error (ac_pc=48; ac_instr_counter=19364690): System Call close (fd 7): Bad file descriptor
make: *** [run] Erreur 1

With multi_16 I have another error than the bad file descriptor, this is this error :
*** Error in `./platform.noc.at.x': corrupted double-linked list: 0x000000000119ee90 ***
make: *** [run] Abandon (core dump créé)

I see some warnings at the simulation beginning about systemc kernel with /sc_object_manager.cpp:149. Do you have some ideas where the problem is? I must say that simple applications are working well, I tested fft and dijkstra.

Thank you again for your support.

Best regards,

Erwan

/!\ PS : I just finished to test multi_parallel and it worked! It seems to come especially from the apps am I right ?

Here is the execution output :
erwan@lo-lester-040:~/Documents/Logiciels/MPSoCBench/MPSoCBench$ ./MPSoCBench -p=mips -i=noc.at -n=4 -s=multi_parallel -r
[['mips'], ['4'], ['multi_parallel'], ['noc.at']]
./platform.noc.at.x multi_parallel.mips.x 4

        SystemC 2.3.1-Accellera --- Apr 16 2015 10:39:56
        Copyright (c) 1996-2014 by all Contributors,
        ALL RIGHTS RESERVED

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_0
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_1
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_2
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_3
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_4
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_5
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_6
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

Warning: (W505) object already exists: noc.thread_node. Latter declaration will be renamed to noc.thread_node_7
In file: ../../../../src/sysc/kernel/sc_object_manager.cpp:149

NOC:

3 Lines

3 Columns

7 Peripherals

4 Masters

3 Slaves

2 Inactive Nodes

4 Slave Empty Nodes

3 Master Empty Nodes

9 Wrappers for Masters/Slaves

Printing mesh!

mesh[0][0]: status-> 1  x->0  y->0
mesh[0][1]: status-> 1  x->0  y->1
mesh[0][2]: status-> 1  x->0  y->2
mesh[1][0]: status-> 1  x->1  y->0
mesh[1][1]: status-> 1  x->1  y->1
mesh[1][2]: status-> 1  x->1  y->2
mesh[2][0]: status-> 1  x->2  y->0
mesh[2][1]: status-> 0  x->2  y->1
mesh[2][2]: status-> 0  x->2  y->2
Printing Table of routs!
Number of Active Lines in the table of routs: 7
m[0]: 536870912 0 0
 m[1]: 553648128 0 1
 m[2]: 570425344 0 2
 m[3]: 587202560 1 0
 m[4]: 603979776 1 1
 m[5]: 620756992 1 2
 m[6]: 637534208 2 0
 ArchC: Reading ELF application file: multi_parallel.mips.x


ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

ArchC: -------------------- Starting Simulation --------------------

MPSoCBench: The simulator is prepared.
MPSoCBench: Beggining of time simulation measurement.

Warning: (W536) immediate self-notification ignored as of IEEE 1666-2011: noc.thread_node
In file: ../../../../src/sysc/kernel/sc_process.cpp:361
In process: noc.thread_node @ 100 ns


--------------------------------------------------------------------
-------------------------  MPSoCBench  -----------------------------
----------------Running: multisoftware platform --------------------
--------------sha, stringsearch, dijkstra, basicmath ---------------
--------------------------------------------------------------------
--------------------------------------------------------------------
---------------------- 4 processadores ----------------------------
------------------------ 4 softwares ------------------------------
-------------------- 1 threads per software -----------------------
--------------------------------------------------------------------
---------------- Start running: stringsearch -----------------------
------------------- Start running: dijkstra ------------------------
------------------- Start running: basicmath -----------------------
--------------------- Start running: sha --------------------------

All 4 threads started...

Join Point->3 of 4 processors are not over yet...
Join Point->2 of 4 processors are not over yet...

ArchC: -------------------- Simulation Finished --------------------
Join Point->1 of 4 processors are not over yet...

ArchC: -------------------- Simulation Finished --------------------

ArchC: -------------------- Simulation Finished --------------------

ArchC: -------------------- Simulation Finished --------------------

Info: /OSCI/SystemC: Simulation stopped by user.
ArchC: Simulation statistics
    Times: 157.20 user, 0.09 system, 156.85 real
    Number of instructions executed: 17334136
    Simulation speed: 110.27 K instr/s
cache: IC
Cache statistics:
Read:   miss: 1989877 (11.4797%) hit: 15344067 (88.5203%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 1989877
cache: DC
Cache statistics:
Read:   miss: 0 (0%) hit: 0 (0%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 0
ArchC: Simulation statistics
    Times: 157.20 user, 0.09 system, 156.85 real
    Number of instructions executed: 20487019
    Simulation speed: 130.32 K instr/s
cache: IC
Cache statistics:
Read:   miss: 134591 (0.656958%) hit: 20352404 (99.343%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 134591
cache: DC
Cache statistics:
Read:   miss: 0 (0%) hit: 0 (0%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 0
ArchC: Simulation statistics
    Times: 157.20 user, 0.09 system, 156.85 real
    Number of instructions executed: 14072635
    Simulation speed: 89.52 K instr/s
cache: IC
Cache statistics:
Read:   miss: 1894229 (13.4614%) hit: 12177362 (86.5386%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 1894229
cache: DC
Cache statistics:
Read:   miss: 0 (0%) hit: 0 (0%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 0
ArchC: Simulation statistics
    Times: 157.20 user, 0.09 system, 156.85 real
    Number of instructions executed: 20408699
    Simulation speed: 129.83 K instr/s
cache: IC
Cache statistics:
Read:   miss: 11499 (0.0563437%) hit: 20397181 (99.9436%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 11499
cache: DC
Cache statistics:
Read:   miss: 0 (0%) hit: 0 (0%)
Write:  miss: 0 (0%) hit: 0 (0%)
Number of block evictions: 0

Total Time Taken (seconds):    156.853116
Simulation advance (seconds):    3.905916
MPSoCBench: Ending the time simulation measurement.

Total Number of Packages:    34831249
Total Number of Hops:    203938372
Average Number of Hops per package:    5
Memory Reads:    28581062
Memory Writes:    6224673

TESTING RESULTS

 SHA Passed.

 Stringsearch Passed.

 Basicmath Passed.

 Dijkstra Passed.

[Erwan Moréac]

Hi again,

Little edit, the application multi_security is working well too as multi_parallel

Best regards,

Erwan

[Liana Duenha]

Hello Erwan,

Related to the SystemC warnings:  The latest version of SystemC generates warnings because its object manager does not accept repeated names for internal components (sc_objects). Then it assigns a number at the end of the name to identify each object solely. This problem occurs only for the NoC-AT, since each tlm_node has an SC_THREAD  (named as  thread_node) and the class tlm_noc declares a mesh as an array of tlm_nodes. 

I've read about it in some SystemC forums and I believe that we can solve this problem using a sc_vector object to define the  mesh (array of nodes).

So, although I have the idea of how to avoid those warnings, I could not spend enough time to solve the problem.

Related to the multi_parallel simulation: I realized that the multi_parallel report shows 0% of hits and misses (for DC cache). These are strange numbers, do you agree?

I simulated multi_parallel on quad-core mips plataforms and the results for DC caches are in bold at the following report:

cache: DC

Cache statistics:

Read:   miss: 268105 (2.73992%) hit: 9517022 (97.2601%)

Write:  miss: 308933 (8.49291%) hit: 3328606 (91.5071%)

Number of block evictions: 370368

ArchC: Simulation statistics

    Times: 157.30 user, 0.27 system, 157.40 real

    Number of instructions executed: 69958764

    Simulation speed: 444.75 K instr/s

cache: IC

Cache statistics:

Read:   miss: 1811017 (2.58869%) hit: 68147709 (97.4113%)

Write:  miss: 0 (0%) hit: 0 (0%)

Number of block evictions: 1734803

cache: DC

Cache statistics:

Read:   miss: 242003 (1.38925%) hit: 17177628 (98.6107%)

Write:  miss: 134157 (10.9044%) hit: 1096139 (89.0956%)

Number of block evictions: 179453

ArchC: Simulation statistics

    Times: 157.30 user, 0.27 system, 157.40 real

    Number of instructions executed: 87599196

    Simulation speed: 556.89 K instr/s

cache: IC

Cache statistics:

Read:   miss: 16668 (0.0190276%) hit: 87582504 (99.981%)

Write:  miss: 0 (0%) hit: 0 (0%)

Number of block evictions: 12512

cache: DC

Cache statistics:

Read:   miss: 5786 (0.0236669%) hit: 24441849 (99.9763%)

Write:  miss: 2621 (0.0767565%) hit: 3412074 (99.9232%)

Number of block evictions: 3130

ArchC: Simulation statistics

    Times: 157.30 user, 0.27 system, 157.40 real

    Number of instructions executed: 31683436

    Simulation speed: 201.42 K instr/s

cache: IC

Cache statistics:

Read:   miss: 1229254 (3.87994%) hit: 30453047 (96.1201%)

Write:  miss: 0 (0%) hit: 0 (0%)

Number of block evictions: 1118221

cache: DC

Cache statistics:

Read:   miss: 314125 (3.45347%) hit: 8781786 (96.5465%)

Write:  miss: 102921 (4.4323%) hit: 2219144 (95.5677%)

Number of block evictions: 207267

IC and DC caches are defined in the processors/mips/mips_nonblock.ac (or mips_block.ac for router and noc-lt platforms).

Example: 

ac_icache   IC("2w", 64, 8, "wt", "fifo");

ac_dcache   DC("2w", 64, 8, "wt", "fifo");

...

IC.bindTo(MEM);

DC.bindTo(MEM);

As your report contains DC and IC statistics (even if 0%) I believe that both caches are defined properly, but still there is something to be solved.

First I will check if the submodules (mips, arm, sparc and powerpc) are out-of-date at the official repository and if there are ArchC commits about this issues. 

I'll let you know when there is a solution.

Liana.

[Erwan Moréac]

Hello Liana,

If systemC warnings are not the source of the current problem, these warnings are not a priority of course.

Indeed I agree for hits and misses 0%, I didn't noticed these numbers. I just wanted to see if MPSoCBench finishes simulations or not to be honest... So the ArchC could be finally the problem of these statistical errors.

I hope you will find a way to solve this. Thank you again.

Best regards,

Erwan

PS : This question doesn't concern the current issue, I would like to know if you have an idea of among all these applications, which one generates the highest traffic during simulation?

[Liana Duenha]

Erwan,

I believe you can find the values you need in my thesis attached. Section 3.2 describes application (limitations and parallelizing techniques) and Section 4.2 show network traffic

results.

:-)

Liana. 

[Erwan Moréac]

Thank you very much Liana! This was exactly what I was looking for.
In fact, I realized in my previous experiments that the NoC is really under-used with some individuals applications. I mean the network can support a much higher traffic that generated by dijkstra for example.

By the way, I will read this part to find the highest traffic. Thank you again.

Good luck for the miss hits problem :)

Best regards,

Erwan